REGIME TYPE: DOES IT REALLY MATTER? On the Perils of Semi-Presidentialism for

Party System Institutionalization in Post-communist Eastern Europe

Abstract

Among political scientists the debate about the relative virtues/vices of the

different regime types has constituted one of the most incandescent fields of inquiry.

Yet few studies address the impact of semi-presidentialism on the prospects for party

system institutionalization, otherwise a necessary, but not sufficient, condition for the

consolidation of democracy (Mainwaring and Scully, 1995; Morlino, 1998). Seeking to

begin to fill this gap in the literature, and departing from a new operationalization of the

party system institutionalization (Casal Bértoa and Enyedi, 2010), this paper constitutes

an attempt to analyse the effects parliamentarism and/or semi-presidentialism produce

on the level of systemic institutionalization in new 17 post-communist Eastern

European democracies. Examining differences in these cases, to which an in-depth

analysis of four representative case-studies is added, the paper finds substantial

evidence that, in clear contrast to parliamentarism, semi-presidentialism is detrimental

to the process of institutionalization in young party systems.

Keywords: party system institutionalization; type of regime; post-communist Eastern

Europe; electoral fragmentation; party institutionalization

Introduction

Among political scientists the debate about the relative virtues/vices of the

different regime types has constituted one of the most incandescent fields of inquiry.

Intellectually, such debate was launched almost two decades ago by the conviction that

institutions matter as they are considered to have an autonomous impact on the

development of party politics. In this context, Linz´s seminal articles on The Perils of

Presidentialism and The Virtues of Parliamentarism, published in 1990, constituted the

milestone sparking much of the subsequent discussion. Since then many scholars across

the world have put their efforts into investigating the relationship between type of

regime and democratic consolidation in new post-authoritarian states. In terms of

regime-related outcomes, however, and in comparison to the work on presidentialism

and parliamentarism, there has been “far less research on the institutional […] outcomes

of premier-presidential [meaning semi-presidential] regimes” (Roper, 2002:263; see

also Elgie, 2004). Moreover, even if party system institutionalization is considered to be

a necessary, but not sufficient, condition for the good functioning of democracy (Kuenzi

and Lambright, 2001; Mainwaring and Scully, 1995; Morlino, 1998; Stockton, 2001),

little has been said about the effect the type of regime may have on the process of party

system institutionalization in new democracies.

Seeking to begin to fill this gap in the literature, this paper attempts to analyse

the effect of the type of regime on the process of party system institutionalization in

new post-communist democracies. In this context, the paper proceeds as follows.

Section one offers an analytical perspective on the conceptualization of semi-

presidentialism and party system institutionalization. Some of the scholarly debate of

the concept is briefly addressed, showing how certain aspects of these two notions have

led to certain confusion and misunderstanding within the literature. The ambition is to

clear out some of those difficulties, and provide concise definitions which enable us to

analyze the relationship between the abovementioned phenomena in new democracies.

Section two summarizes some of the most important arguments advanced by scholars in

the course of the regime type debate, making several new propositions concerning the

possible implications the type of regime may have for the institutionalization of party

systems while testing their validity on the basis of the empirical evidence from 17 post-

Eastern European democracies. Finally, section three analyses the causal mechanisms

linking type of regime and party system institutionalization in four representative

countries.

Semi-presidentialism and Party System Institutionalization:

Two Debated Concepts

As it follows from the substantial body of recent literature devoted to the

concept (Bahro et al., 1998; Elgie, 1998, 1999; Pasquino, 1997; Roper, 2002; Sartori,

1997; Siaroff, 2003), the notion of semi-presidentialism has been particularly prone to

definitional problems. Different scholars have adopted or launched quite different

definitions of the concept (O´Neill, 1993; Blondel and Müller-Rommel, 2001; Sartori,

1997; Linz, 1994) and, as a consequence, the identification of the set of semi-

presidential countries has varied from writer to writer (Elgie, 1998, 1999; Sartori,

1997:122).

In the current paper, I will rely on Elgie´s purely constitutional definition of the

concept, by now the most common way of defining this type of regime (e.g. Elgie 2005;

Elgie and McMeniman, 2008; Elgie and Moestrup, 2007, 2008; Kirschke, 2007;

Schleiter and Morgan-Jones, 2006; Protsyk, 2005a; Skach 2005; Shugart 2005; 2006).

Thus, semi-presidentialism is considered to be “a regime where there is both a popularly

elected fixed-term president and a prime minister and a cabinet responsible to the

legislature” (1999:13).

In contrast to others´ (Duverger, 1980; Pasquino, 1997; Sartori, 1997), this is a

purely constitutional definition which has the advantage of being (1) minimal, as it

focuses solely on the mode of election and removal of the two heads of the executive;

(2) precise because, omitting any reference to the powers of the president, it minimises

the opportunity for variation in case selection from one writer to the next; and (3)

parsimonious because simply relying on what is considered to be the same basic

constitutional structure, it allows for a more accurate differentiation between semi-

presidentialism and all versions of parliamentarism and presidentialism (Elgie and

Moestrup, 2007).1

Similarly, and although it may be difficult to believe given its central

importance, the concept of party system institutionalization has no established

definition. Putting it very briefly, and summarizing a discussion sketched out in an

earlier paper (Casal Bértoa, forthcoming), most authors dealing with the concept simple

1 This is not to say, however, that it does not pose any problems, as Elgie himself has recognised (2008:51). In fact, one of the most important borderline cases is Slovakia, where the president can be removed from office by a plebiscite. However, as Elgie (2005) has argued, the removal of a president by plebiscite cannot be equated to the removal of the head of state by a vote of confidence in the assembly. In this sense, the Slovak president should be considered to occupy a fixed-term presidency.

propose a series of “dimensions” of the notion (Morlino, 1998; Bielasiak, 2002;

Grzymała-Busse, 2002; Meleshevich, 2007), without paying much attention to the

conceptualization itself.

Notwithstanding the latter, and despite the fact that no two scholars have arrived

at the same final combination of dimensions of institutionalization, the truth is that all

meanings of the conception of party system institutionalization contain the idea of

stability and persistence in the rules and nature of inter-party competition (Lindberg,

2007; Mainwaring and Scully, 1995; Mair, 2001; Przeworski, 1975). As a consequence,

and bearing in mind that the core of a party system is to be found in the patterns of

interaction among its subunits (i.e. political parties; see Sartori, 1976; Mair, 2006), I

consider party system institutionalization to be the process by which the patterns of

interaction among political parties become routine, predictable and stable over time

(Bakke and Sitter, 2005; Mair, 2001).2 In other words, a system of parties can be said to

be institutionalized when political parties cooperate, collaborate and colligate in a

standardized and structured way - a way that is independent of the relevant issues in

each moment and which random shocks cannot alter (Mainwaring, 1998).

In order to assess the level of institutionalization in new East Central European

party systems, I will rely on Mair´s (1996, 2007) framework for party system analysis

which, focusing on the patterns of inter-party competition for government, enables to

determine whether a party system is or not institutionalized (table 1). Putting it briefly,

party systems are considered to be institutionalized if (1) alternations of governments

are either total or none, (2) governing alternatives are stable over a long period of time,

and (3) some parties (“outsiders”) are permanently excluded from participation in

national government and weakly institutionalized when there are (1) partial alternations

of governments, (2) no stable compositions of governing alternatives and (3) access to

government has been granted to all relevant parties.3

In order to minimize subjective judgements and opinions in the measurement of

the elements of party system institutionalization, I quantitatively operationalize each of

the factors suggested by Mair (Table 1). First of all, the degree of governmental

alternation is measured by a so-called index of government alternation (IGA – see Mair

2 “Interaction” refers to the creation and dissolution of coalitions and alliances, or changing between behaviours of cooperation and competition. 3 It should be borne in mind that, notwithstanding these two limiting cases, the degree of closure varies, ranging along a continuum from situations where it is more pronounced to those in which it is less, thereupon reflecting different degrees of institutionalization.

2007:140), which simply adapts Pedersen´s (1979) well-known index of electoral

volatility to the measurement of ministerial volatility.4 The second criterion, based on

assessing whether or not the party or combination of parties has governed before in that

particular format, is captured by an index of familiar alternation (IFA), which measures

the percentage of ministries belonging to familiar combinations of parties.5 Thirdly,

access to government is measured by the index of entry closure (IEC), which basically

calculates the percentage of ministers belonging to “old” governing parties.

Table 1. Criteria and Operational Indicators of Party System Institutionalization Dimension Criteria Indicators Operationalization

Government Alternation

IGA: Pedersen´s index of ministerial volatility (MV)

. If MV≥50, then IGA = MV

. If MV<50, then IGA = 100-MV

Governing Formulae

IFA: % ministers of all familiar governments

. If the very same combination = 100%

. If entirely new combination or new party forms single party government = 0% . If part of the new government is familiar = % of the familiar part . If a party earlier in government forms a government on its own = 100 - % of previous coalition partners

Stability

Access to government

IEC: % ministers from “old” governing parties

“Old” governing parties (see Sikk, 2005)

Source: Adapted from Casal Bértoa and Enyedi (2010:19)

Because time is particularly important when trying to measure the level of party

system institutionalization in a country (Mair, 1997), I also take into consideration all

the years a particular cabinet has lasted, understanding that if there have been two or

more cabinets in one year, then the averages of the scores for the different above-cited

government features are considered to characterized the year better than any of such

individual factors (Casal Bértoa and Enyedi, 2010:20). Finally, and in order to avoid

measuring incompatible scores, I use the standardized (z-) score of the three variables.

The addition of all those three z-scores gives the final degree of system

institutionalization in what I have named “composite index of party system

institutionalization” (iPSI) which, giving weight to all the elements of stability, has the

advantage of paying due attention to it as the sole dimension of institutionalization.

An overview of the level of party system institutionalization in new post-

communist party systems is shown in Figure 1, which ranks the 17 Eastern European

4 Ministerial volatility is computed by adding the net change in percentage of “ministers” (including the primer minister), rather than ministries or portfolios, gained and lost by each party in the cabinet from one government to the next, and then dividing by two. 5 For detailed instructions on how to consider a combination of parties familiar, please see Casal Bértoa and Enyedi (2010).

democracies6 in terms of the stability in the structure of inter-party competition for

government during the period 1990-2009. Apart from the pronounced cross-national

variation shown by these summary data, the most evident conclusion to be drawn is that

while the four party systems at the top of the ranking (i.e. Kosovo, Hungary, Albania

and the Czech Republic) are all instances of parliamentarism, the three least

institutionalized party systems (i.e. Ukraine, Poland and Lithuania) are semi-

presidential.

Figure 1. Party system institutionalization in 17 post-communist Eastern European countries7

-10

-8

-6

-4

-2

0

2

4

6

Party System Institutionalization

It is also interesting to note that it is only among semi-presidential regimes that a

prevailing tendency towards a lack of systemic institutionalization can be found. Indeed,

it is by simply looking at the weakly institutionalized party systems (iPSI ≤ 0) that we

can find only two parliamentary regimes (i.e. Latvia and Estonia), while the other six

(i.e. Macedonia, Romania, Bulgaria, Lithuania, Poland and Ukraine) are all instances of

semi-presidentialism. Slovakia, after the adoption of semi-presidentialism, is just on the

verge (iPSI = 0.1). On the other hand, while none of the parliamentary systems can be

considered to be under-institutionalized (iPSI ≤ -1), only two (i.e. Croatia and Slovenia)

out of eleven semi-presidential regimes is considered to be institutionalized (≥ 1).

6 All of the countries included in the current study are considered to be “minimally” democratic, as they score 6 or higher on the polity2 variable from the Polity IV dataset (Polity IV Project 2009). Montenegro has been excluded due to the lack of enough analytical time (at least one year) since the “founding” elections. 7 Moldova and Slovakia are divided into two different periods according to the distinct type of regime adopted.

All in all, the truth is that while party systems in post-communist Eastern Europe

have institutionalized in different ways and at different rates (Casal Bértoa and Mair,

forthcoming), it follows from the data above that parliamentary systems tend to be more

institutionalized than semi-presidential ones. It is to explain this relationship between

systemic institutionalization and type of regime that I will devote the rest of this paper.

On the Perils of Semi-presidentialism for Party System Institutionalization

Although since Linz´s (1990a/b) seminal work a great deal of research has been

devoted to analyse the effects of constitutional regime choice on the consolidation and

functioning of democracy (in general), the truth is that the vast majority of work has

focused both on presidentialism and parliamentarism and, only to a slightly lesser

extent, on semi-presidentialism. Yet, to the extent that scholars have theorized about it,

they have arrived to a rather straightforward conclusion, namely: semi-presidentialism

should definitively be avoided (Colton and Skach, 2005; Lijphart, 2004; Linz, 1994;

Stepan and Skach, 1993; Mainwaring and Shugart, 1997; Suleiman and Stepan, 1995;

Moestrup, 2004; Valenzuela, 2004), especially in young democracies.8

In particular, semi-presidentialism has three major problems, namely: the fixed

term of the head of state makes semi-presidential regimes prone to cohabitation; the

president and the cabinet have competing claims to legitimacy; semi-presidential

regimes are more prone to experience divided minority government (Elgie, 2007, 2008).

The bottom-line argument is then that semi-presidential regimes are less likely to be

conducive to stable democracy than parliamentary systems. In this context, Stepan and

Suleiman speak for a long line of scholars when they note that semi-presidentialism “is

a more risk-prone system than the modern parliamentarism that has evolved in Europe

[…]” (1995:412).

It should be borne in mind, however, that all these criticisms of semi-

presidentialism developed in the context of democratic transition in post-authoritarian

states and were used to explain democratic survival and/or collapse. Still, to the extent

that they can be extrapolated to the analysis of party system institutionalization in new

democracies, they seem to suggest that the higher the power of the president, the lower

the level of systemic institutionalization will be (Mainwaring and Shugart, 1997;

Sartori, 1997; Shugart and Carey, 1992; Skach, 2005, etc.).

8 This is not equivalent to say, however, that semi-presidentialism does not have any friends (e.g. Duverger, 1997; Fish, 2001a; Frison-Roche, 2005; Pasquino, 1997; Sartori, 1997), although they are obviously fewer than in the case of parliamentarism.

Figure 2. Presidential powers and party system institutionalization in post-communist Eastern Europe (1990-2009)

In order to test the previous assumption, figure 2 plots the level of party system

institutionalization in a country with the degree of power post-communist heads of state

have at their disposal (see Spörer, 2004). Although the correlation between the previous

variables is not only significant (at 0.01 level), but also relevant (r = -0.698), one

important observation can be drawn from an in-depth look at the particular cases: once

the two countries with least institutionalized party systems and the most powerful

presidents (i.e. Ukraine and Poland up to 1996) are excluded, the relationship between

the two variables here analyzed becomes insignificant as well as irrelevant (r = -0.277).9

In fact, a quick examination of the cases reveals similar levels of party system

institutionalization in countries where the degree presidential power totally differs (e.g.

Czech Republic and Croatia, Latvia and Bulgaria, or Estonia and Lithuania). On the

contrary, in countries where the power of the president remained unchanged despite the

transformation of regime type, the level of systemic institutionalization clearly

9 When both Ukraine and Poland (either up or from 1996) are excluded the correlation score is as low as -0.055.

fluctuates (e.g. Slovakia). This is not to say, however, that as some scholars have

suggested (see Elgie and Moestrup, 2008) the powers of the president are irrelevant

when trying to explain party system development and institutionalization in semi-

presidential regimes. In fact, in this case the correlation is both significant and relevant

(r = -0.788, sig at 0.01 level), even when the two abovementioned most extreme cases

are excluded (r = -0.567, sig. at 0.05 level; R² = .32).10 However, this is not the aim of

this paper which, as stated at the beginning, focuses on the effects different types of

regime – parliamentarism and semi-presidentialism, in concrete - may have on the

process of party system institutionalization in new post-communist Eastern European

democracies.

Still, what really follows from the above analysis is that, contrary to the large

amount of literature focusing on the powers of the president to explain democratic

consolidation in post-transitional countries, the former cannot be employed to explain

why new post-communist party systems diverge in their level of institutionalization. In

this context, and because of those power-centred approaches scholars have failed to

anticipate, generally speaking, the explosive repercussions the popular election of the

president can pose for the process of party system institutionalization itself.11

Figure 3. Type of regime effects on party system institutionalization

10 Still, the correlation between party system institutionalization and presidential power in parliamentary regimes is not only both insignificant and irrelevant (r = 0.043), but also in the “wrong” direction. 11 The only exception is Bartolini´s (1984) work on the direct election of the head of state in Western Europe, even if he does not directly deal with systemic institutionalization. In this context, even Meleshevich´s (2007) recent study on the relationship between type of government and political institutionalization in Post-soviet countries suffer from the abovementioned power-centric conceptualization.

Party institutionalization

Type of regime

Parliamentary fragmentation

Structure of competition

According to my understanding and as it follows from figure 3, the perils of

semi-presidentialism for the institutionalization of a country´s party system can be of

two different types: indirect and direct.12

Indirect effects: Fragmentation and...

Presidential elections are commonly thought to influence the size of the

legislative party system through a coattails effect (Golder, 2006),13 although it is not

very clearly in the literature if such influence will have a increasing (Filippov,

Ordeshook, and Shvetsova, 1999; Jones, 1995; Linz, 1994), or a reductive character

(Cox 1997; Mozaffar, Scarritt, and Galaich 2003; Rose and Tikhomirov, 1996; Shugart

and Carey, 1992).14 Although all these claims seem to be contradictory, the answer to

this dilemma is to be found – as usual in political science - somehow in the middle,

namely, it will depend on the specific institutional arrangements a country has adopted.

In general, scholars agree that when determining the electoral effects of semi-

presidentialism on the number of parties all of the mechanical and psychological effects

of single-member district elections (Cox, 1997; Duverger, 1954; Sartori, 1997) apply.

At the end of the day, and because presidential elections have a winner-take-all

character (i.e., district magnitude is 1), presidential elections in semi-presidential

regimes should be considered to have a reductive effect.15 However, and according to

the literature on the consequences of electoral systems, such reductive effect will be

different depending on the electoral rule employed. Thus, if the electoral rule is

plurality, the popular election of the head of state will have a reductive effect as, on the

one hand, political elites will tend to form broader coalitions of like-minded parties, and

voters may opt for strategic voting out of fear of wasting their votes (Mainwaring and

Shugart, 1997:36). In Moser´s own words:

12 it is important to note here that, rather than law-like generalizations that inexorably links semi-presidentialism with party system under-institutionalization, each and every of the propositions presented in this section need to be seen as tendencies and arrangements that “can”, but may not, hinder institutionalization. In other words, it is a question of potentiality, as each of the regime-related effects on the process of systemic institutionalization proposed below will depend on both institutional arrangements (i.e. timing, sequence and electoral system) and, obviously, personal agency (i.e. coalition bargaining). 13 Although the major part of the research on the effects of presidential elections on the “effective” number of parties derives from the “presidentialist” literature (e.g. Jones, 1995; Mainwaring, 1993; Mainwaring and Shugart, 1997; Shugart and Carey, 1992), I consider it to be also relevant for the purposes of this analysis, as in semi-presidential regimes the head of state is also popularly elected. 14 To complicate the debate, some scholars state that the popular presidential elections may have no effect at all (e.g. Samuels 2000). 15 In fact, in the majority of new European democratic countries the “effective” number of presidential parties has often been lower than both the actual and effective number of electoral parties.

The fear of splitting the vote within a specific ideological camp's potential

electorate and allowing victory to a candidate from the opposite end of the

political spectrum further reinforces impulses for consolidation [meaning

concentration] (2001:98)

On the contrary, if the electoral formula is majoritarian (i.e. two-round),

presidential elections will have the opposite effect. The main reasons for that are three,

namely: (1) because presidential elections are the unrivalled event in the political

calendar of any country,16 competition for the presidency enhances the public visibility

of politicians through increased media exposure and campaign contributions, boosting

their options in future parliamentary elections (Filippov, Ordeshook, and Shvetsova,

1999; Müller, 1999:42-43; Gallagher, 1999:113); (2) “the expectation of a runoff

increases the incentive to compete in the first run, either in the hope of placing among

the two most favoured or of gaining bargaining power for support in the runoff of one

of the two leading contenders” (Linz, 1994:22; see also Golder, 2006:42, or Wright and

Riker, 1989);17 (3) last but not least, because presidents – even if not powerful - are

usually seen as the leader of the nation by the majority of the population, the weight of

the presidency encourages the formation of parties by ruthless politicians in order to

simply satisfy their “personal” ambitions (e.g. Paksas in Lithuania, Eanes in Portugal,

Snegur in Moldova, etc.).

Likewise, the extent to which presidential elections will have a reductive effect

on the number of parties will also depend on the electoral cycle, concretely, on the

temporal proximity of presidential and legislative elections. Thus, and due to the

“spillover” effect that the presidency has on the behaviour of voters and party elites in

legislative elections (Mainwaring, 1993), if both types of elections are held

concurrently, presidential elections will have a reductive effect on the effective number

of electoral parties. Conversely, if the two elections do not coincide, such reductive

effect will be definitively weaker (Jones, 1995; Lijphart, 1994; Mainwaring and

Shugart, 1997; Amorin Neto and Cox, 1997; Sedelius, 2006; Shugart and Carey,

1992).18

16The fact that the electoral turnout in presidential elections has been, generally speaking, significantly higher than in parliamentary elections seems to confirm, beyond all doubt, this assumption. 17 In fact, as Golder (2006:47) has maintained, the adoption of run-off provisions will only “exacerbate any legislative fragmentation caused by the use of proportional representation in legislative elections”. 18 For an opposite argument see Filippov, Ordeshook, and Shvetsova (1999), although they themselves recognize the (statistical and temporal) limitations of their analysis.

All in all, and taking into consideration all what has been said, I may conclude

stating that popular elected presidents will increase multi-partism in those systems

where presidential elections does not have a concurrent/plurality character.

Figure 4. Type of regime and the “effective” number of electoral parties in post-communist Eastern Europe (1990-2009)19

0

2

4

6

8

10

12

14

16

18

20

ENEP

Figure 4 ranks all post-communist party systems according to the “effective”

number of electoral parties (ENEP), calculated according to Laakso and Taagepera´s

(1979) classic index. As is evident from the figure, Ukraine reflects by far the most

pronounced levels of electoral fragmentation, followed some way back by Estonia (in

1992) and Slovakia (from 2002). At the other side of the ranking (i.e. among the most

electorally concentrated party systems), Moldova (from 2001) and Albania (both

parliamentary regimes), are found. Moreover, and with the only exception of Latvia,

among the most electorally fragmented party systems (ENEP ≥ 6) can be found,

namely: Poland, Lithuania and Slovenia, on top of the above-cited Slovakia (2002-

2009), Estonia (1992) and Ukraine.

Interestingly enough, when both variables are statistically correlated,20 the

resultant coefficient (-0.260), although in the expected direction, seems to deny any

relevant relationship between them. Still, if a plot (not included) of both variables is

19 Although Estonia is generally considered to be a parliamentary regime, the first Estonian presidential elections had a direct character and, therefore, need to be considered among the semi-presidential ones, at least in terms of the ENEP. The reason for that is that the first democratic elections (both presidential and parliamentary) since the restoration of democracy in Estonia were concurrently held on September the 20th, 1992. 20 Type of regime (ToR) is operationalized as a dichotomous variable, that is: parliamentarism (1) and semi-presidentialism (0).

undertaken, it is possible to distinguish two rather clear outliers, one among each of the

parliamentary and semi-presidential countries, namely: Latvia and Ukraine,

respectively. When those two systems are excluded, the correlation coefficient increases

up to -.327, revealing a certain negative relationship between electoral fragmentation

and type of regime worthy of study. All in all, and as it follows from table A in the

Appendix, the truth is that post-communist Eastern European semi-presidential regimes,

with an average ENEP of 6.7, tend to be more electorally fragmented than

parliamentary systems (ENEP = 5.2).

... Party Institutionalization

In general, it has been argued that semi-presidentialism has acted as a major

impediment to the development of institutionalized political parties (Colton, 1995; Fish,

2001b; Huskey, 1997).21 Scholars have cited several reasons as to why this has been so.

First, contrary to parliamentary presidents who, not being elected by popular

vote, have a strong interest in party building, the institute of a popularly elected

president encourages greater personalism, making it more difficult for parties to develop

coherent programmes and identities (Mainwaring, 1993; Moser, 1998; White, Wyman

& Kryshtanovskaya, 1995). Moreover, because – as we have already seen – presidential

candidates need to seek a broader mandate than any given party, the popular election of

the president often induces “[c]ampaign personalization [and] reduces the relevance of

party platforms and party organization” (Samuels, 2002:480). At worst the popular

election of the head of state “may serve as an incentive to demagoguery and populism”

(Mainwaring and Shugart, 1997:32; see also Harmel and Janda, 1982). In this sense,

semi-presidentialism may even invigorate the danger of bonapartism, characterized by

the instrumentalization of political parties with obvious populist traits (Bahro et al.,

1998:217).

Second, the personalized character of a presidential race provides no safeguard

and not buffer against political “outsiders”, with little or no political experience, seeking

election (Elgie, 2001; Linz, 1994; Stepan and Suleiman, 1995). As Paksas´ victory in

the 2002 presidential contests in Lithuania shows, these individuals may create parties

at the last minute in order to run for the presidency, therefore, finding it very difficult to

develop parties with strong linkages in society. Moreover, when successful, such

presidents tend to ignore their own political parties, personalizing the partisan process

21 For a critique of this hypothesized relationship, see Ishiyama and Kennedy (2001).

(Stepan and Skach, 1993:20). Thus, semi-presidential regimes are considered to

contribute to “the creation of small and ephemeral parties, most often the personal

vehicles of presidential candidates and little more” (Cadoux, 2007:96). This clearly

contrasts with presidents in parliamentary systems, usually long-term career politicians

and, in many cases, also former party leaders (either in power or in opposition) over

many years. For them, political parties are clearly infused with value.

Third, while the incentive structure in parliamentary regimes encourages party

discipline and, therefore, institutionalization of party organization, semi-presidential

regimes have no such incentives for party loyalty (Epstein, 1967; Linz, 1994; Moser,

2001). In this sense, semi-presidentialism may contribute to factionalism, that is, to the

institutionalization of division within parties rather than between parties (Azebedo and

Nijzink, 2007), leading in the most acute case to the break-up of those political

organizations.

A final criticism of semi-presidential regimes results from the “above party-

politics” character of presidents, which in itself often appears to be a desideratum.

Contrary to parliamentary regimes where, by definition, the support of a parliamentary

party is a must for a successful candidate to become president, semi-presidentialism

encourages the image of a president who is above and against political parties (Huskey,

2007; Linz, 1994; Meleshevich, 2007). Certainly, a non-party president which portrays

him/herself as a representative of the whole nation22 will obviously hinder the process

of party institutionalization, as individuals will “focus on forming personal attachments

with presidential hopefuls, bypassing association with political parties” (Ishiyama,

2008:42; see also Meleshevich, 2007).23 In sum, it seems to be a commonly held notion

among scholars that semi-presidentialism is the culprit for the lack of cohesive,

disciplined, programmatic, socially rooted political parties in a country.

As can be seen in figure 5 below, which employs Lewis´ (2006) “index of party

stabilization” (ips),24 political parties tend to be more institutionalized in parliamentary

22 In parliamentary republics, even when elected with the support of a particular party or coalition, heads of state tend to adopt a neutral, rather than negative, attitude towards parties. 23 In this respect, the constitutional provision in Bulgaria, Romania or Lithuania that the president should not be a member of a party is in clear contradiction with a prerequisite for effective party development and institutionalization (Filippov, Ordeshook, and Shvetsova, 1999:19). 24 The “ips” involves the progressive enhancement of the proportion of the total vote for political parties in a given election over time - by 20% for a party’s second appearance in parliament, 40% for the third, 60% for the fourth, 80% for the fifth, and so forth (Lewis, 2006:574-575). The logic is that, taking notice of both voter stability (in voters´ electoral preferences) and the age of a party organization, this index measures the two most important dimensions of party institutionalization, namely: stable roots in society and party organization (Casal Bértoa, 2010).

systems than in semi-presidential ones.25 In fact, while among the nine systems with a

higher degree of party institutionalization six are parliamentary, eight out of the nine

systems where political parties are least institutionalized are semi-presidential:

parliamentary Latvia constitutes the only exception. Hence, and as it follows from the

third column in table A, the truth is that political parties tend to be more

institutionalized, both in terms of rootedness and systemness, in parliamentary regimes

(ips = 84.7) than in semi-presidential ones (ips = 75.6). In conclusion, it is possible to

conclude that in post-communist Eastern Europe there is a positive relationship between

type of regime and party institutionalization (r = 0.415).

Figure 5. Type of regime and party institutionalization in post-communist Eastern Europe (1990-2009)

0

10

20

30

40

50

60

70

80

90

100

Party Institutionalization (ips)

Direct (Negative) Effect on the Stability of the Patterns of Inter-party Competition

As it results from Elgie´s definition, a critical departure of semi-presidentialism

from parliamentarism is the presence of a “double electoral process” (Bartolini,

1984:227) which introduces a potential for instability in the patterns of inter-party

competition certainly absent in the latter.

Contrary to parliamentary presidents, which tend to be elected either as the fruit

of a compromise between the totality/majority of political parties or by a qualified

majority which forcefully requires the support of the major parliamentary parties,26

25 Kosovo has been excluded from the analysis here, as it does not fulfill Lewis´ “two legislative elections” requirement (2006:571). 26 Parliamentary political parties tend to see this “almost compulsory” collaboration as ad hoc and strictly occasional.

presidential candidates in semi-presidential regimes face a different and broader

“electoral” constituency (i.e. the electorate as a whole). Because they cannot afford to

ignore any more or less significant segment of the population, broad coalitions which

may include not only extremist political organizations but also any other political

parties with a minimum electoral strength are likely to be formed. The main implication

of this might be that, as a reward for their support in elections, “one or more of them

can plausibly claim to represent the decisive electoral bloc in a close contest and may

make demands accordingly” (Linz, 1990a:58), namely: participation in government,

future electoral cooperation, etc. In this context, an element of instability in the structure

of inter-party competition is introduced as the majority of leading presidential

candidates will be “forced” to seek the cooperation of forces with which they would

otherwise not be ready to collaborate, giving way to previously unseen cabinet

coalitions or to the inclusion of a party excluded from government up to that time. This

is clear in single-round elections, where electoral plurality might depend on even the

small number of voters those “unusual partners” might be able to provide, but most

especially in two-round elections, where political alliances come closer to the process of

coalition formation in a parliament in search of a prime minister.27

Moreover, because “the [presidential] ʻmajority̒ generated might not represent

a politically more or less homogeneous electorate or a real coalition of parties” (Linz,

1994:21), these alliances of “presidential” parties tend to be very fragile and short-

termed, since ideological and policy differences among heterogeneous member-parties

of a loose (pre-)electoral presidential coalition are likely to broaden with time.

Therefore, while in parliamentary regimes parties occupying different electoral niches

normally compete against one another and the coalition of ideologically “close” parties

generally takes place after the election and are binding; in semi-presidential regimes, the

majoritarian character of the presidential elections not only may change the existing

structure of inter-party competition, but tends also to impregnate the new patterns of

interaction with a loose and temporary character.

A second way in which party system stability can be hampered by the mode of

the election of the head of state derives from the higher status of the latter in semi-

27 Clear examples of how the mode of the election of the president modifies the existing structure of inter-party competition are Finland (1937), Bulgaria (2001), Niger (1999), Croatia (2000), Guinea-Bissau (2005), East Timor (2007) or, most recently, Slovenia (2008). In all these cases, the composition of the governmental coalition after parliamentary elections was determined by the patterns of inter-party collaboration during previous presidential elections, reducing the degree of party system institutionalization to a certain extent.

presidential regimes as compared to the position of the monarch or the president in

parliamentary countries. In fact, even if the formal prerogatives of popular elected

presidents are rarely much greater than those of parliamentary heads of state – notable

exception are that of Russia, Ukraine and other post-Soviet states - they have often been

able to use their greater “political legitimacy” prestige to stretch their influence in the

making and breaking of governments and/or parliamentary coalitions. This is not to say,

however, that in some parliamentary regimes the head of state cannot play an active in

the formation of governments and/or the interaction of political parties. However,

because the president or the monarch is not independent of the assembly, he/she cannot

claim to be “tribunes” above the partisan fray (Baylis, 2007:89). Thus, while in Italy,

Germany or even Spain, the head of state will tend to follow the existing patterns of

interaction among political parties, strengthening, rather than weakening, the stability of

the system of competition as a whole; in semi-presidential regimes, the popular election

of the head of state is likely to imbue presidents with a feeling that they, “although

initially the choice of a small proportion of the electorate, […] represent a ̒true and

plebiscitary̒ majority” (Linz, 1994:21) and that, consequently, they do not need to

respect the “usual” process of coalition formation and partisan cooperation, introducing

a potential for unpredictability and instability (see also Bartolini, 1984; and Duverger,

1980) which, on the other hand, will be higher the “fresher” the presidential mandate

(Protsyk, 2005b:737; Shin-Goo, 2004). The logic is that the most recent electoral

legitimacy of the president provides him/her with additional political leverage in the

process of government formation, even in the case when, like in Bulgaria or in Ireland,

his formal powers are severely restricted.

Finally, it is not only the sequence, but also the timing of presidential and

parliamentary elections in semi-presidential regimes that enhances the likelihood of

party system instability. Thus, as several scholars have found, presidential coattails will

be higher the closer the two elections are held (Cox, 1997:209-219; Sedelius, 2006;

Shugart and Carey, 1992:226-258).28

As we had the opportunity to see in section 1, there is a positive relationship

between the type of regime and party system institutionalization (r = 0.385): on average,

parliamentary systems (iPSI = 1.2) are more institutionalized than semi-presidential

28 Romania, where both presidential and legislative elections have been traditionally held simultaneously since 1990, constitutes the most evident example.

ones (iPSI = -0.9). Moreover, and as it follows from table A, the latter tend to remain

weakly institutionalized (iPSI ≤ 0), in clear contrast to the former (iPSI > 0).

Table 2. Regime-related outcomes for party system institutionalization Direct effects on… Indirect effects on… Type

of regime Structure of competition PI Electoral fragmentation (ENEP) Parliamentarism n/a 0/+ + n/a n/a +

Semi-presidentialism - 0 - + - - Electoral timing Concurrent Non-concurrent N/A Concurrent Non-concurrent

Electoral sequence

Presidential-Parliamentary

Parliamentary-Presidential

Electoral system

Plurality Majority (run-off)

Plurality

Notes: “+” = positive effect; “-” = negative effect; “0” = non-effect; “n/a” = not available.

Table 2 summarizes the effects different regime types may have on the process

of party system institutionalization in new democracies. In general, parliamentary

regimes will be “institutionalization-friendly”, as they may have a positive impact or, at

maximum, a non-effect on the process of systemic institutionalization, although never a

negative one. In contrast, semi-presidentialism will always have a negative impact on

the stabilization of the structure of inter-party competition, but in two instances: (1)

when both types of elections are not held concurrently and the parliamentary contests

precede to presidential ones; (2) last but not least, in those cases when both elections are

held concurrently, presidential elections will have a reductive effect on the effective

number of electoral parties, but only if a plurality rule is applied.

Party System Institutionalization and the Type of Regime in

Four Selected Post-communist Eastern European countries

As it follows from the previous section, which analyses the impact that on party

system institutionalization may have different regime types in post-communist Eastern

Europe, semi-presidentialism - in clear contrast to parliamentarism - has, in general, a

detrimental effect (either directly or indirectly) on the process of systemic

institutionalization. However, and because correlation is not causation, in this section I

complement the previous analyses with a procedure particularly suitable to make out

“the intervening causal process - the causal chain and causal mechanism - between an

independent variable […] and the outcome of the dependent variable” (George and

Bennett, 2005:296).

The idea is that, by breaking down the rather large process of systemic

institutionalization into its constituent mechanisms, I can more easily identify the “chain

of causation” leading from type of regime to party system institutionalization. With

such aim in mind, and using “process-tracing”, I will proceed next to analyze the

specific relationship the abovementioned variables may have in four different, although

very representative, cases: one is a paradigm of systemic institutionalization (Hungary),

the other constitutes a clear instance of weak institutionalization (Poland), while the last

two are examples of “type of regime” change, either from parliamentarism to semi-

presidentialism (Slovakia) or vice versa (Moldova). Let´s examine then each of them in

turn.

Hungary

In Hungary the head of state is elected by the Országgÿlés (i.e. parliament) for a

five-year term, with the possibility of just one re-election. Because any presidential

candidature mandatorily needs the support of at least 50 parliamentary members, larger

legislative parties tend to control the nomination process. Moreover, the nominee of the

governing coalition at the time of the presidential selection is clearly favoured by the

rules according to which the head of state is appointed: in case a qualified majority of

two-thirds (of all MPs) is not obtained in the first two rounds, a third and last round is

held where a simple majority - of all votes cast in a contest between the two-front

runners - is enough.

On 3 August 1990, Árpád Göncz, a well-known writer and dissident who had

been active during the 1956 revolution, became the first democratically elected

president of Hungary as a fruit of a compromise between the two main political parties

at the time, namely: MSZP and SZDSZ. Contrarily, the next three presidential elections

departed from this consensual style, clearly responding to the existing patterns of inter-

party competition (see Appendix).

Thus, both the 1995 and 2000 election of the president was the maximum

expression of the confrontation between the government and the opposition. The main

difference was that, while President Göncz was re-elected for another five year term in

1995 with the sole votes of MSZP-SZDSZ (O´Neil, 1997:215), in 2000 Ferecn Mádl, a

professor of law not officially affiliated with any party and President Göncz´s rival in

1995, was elected on the third and final round with the “sole” support of both Fidesz

and FKGP, as he had failed to win the supermajority required in the first two rounds of

voting (Tavits, 2009:166).

The 2005 presidential election marked a slight turn in this trend as for the first

time the governing coalition did not manage to nominate a consensus candidate. Still, as

in the previous two elections, a candidate from the government was pitted against a

candidate from the opposition. Katalin Szili, president of the National Assembly and

former vice-president of the Socialists was nominated by the MSZP. Because old

President Mádl decided not to run for re-election, László Sólyom, a former president of

the Constitutional Court and not an active politician, became the nominee of the

opposition parties (Fidesz and MDF). In protest for the Socialist unilateral decision, the

junior coalition party (SZDSZ) refused to participate in all rounds of the vote. In the

event, Sólyom won the election in the third round by a three vote margin (185 to 187).

Although the SZDSZ had not given up their position and, consequently, had not voted

for Szili, this conflict between the governing parties did not threaten the coalition

agreement (Ilonszki and Kurtán, 2006:1126), as the 2006 parliamentary elections clearly

demonstrated (table B).

Hence, Hungarian presidential contests confirm my earlier expectations about

indirect elections, namely, that they are decided either by compromise (e.g. Gönz´s) or

by the pre-existing structure of partisan contestation (e.g. the rest), especially if – as it is

the case – the sequence of elections respond to the parliamentary-presidential pattern

(see Appendix).

On the other hand, while inexistent in terms of fractionalization, the effect of the

indirect mode of election of the president on the process of party institutionalization has

been quite positive. In this context, both the previously mentioned electoral sequence

(i.e. parliamentary-presidential) and, mainly, the above-cited normative on the selection

of the head of state has helped the party in public office to dominate the process of

presidential nomination and appointment and, therefore, consolidate its “grid” on the

party in central office (van Biezen, 2003), definitively institutionalizing its central

position within their membership as well as the electorate as a whole.29

Poland

The Polish president is directly chosen by the electorate for a five-year term,

with the possibility of one re-election. The electoral system used is a majority runoff so,

if no candidate receives at least 50 percent of the vote in the first round, a second round

between the two front-runners takes place. The final winner is appointed head of state.

In Poland, the popular election of the president has had a very negative effect on

the process of party system institutionalization since the very beginning, as it has

29 Although some might see the emergence of István Csurka´s Party of Hungarian Justice and Life (MIÉP) out of the MDF in June 1993 as a negative influence on the process of party institutionalization, the truth is that, when looking in depth at the way in which such split took place, it seems obvious that it was more influenced by personal and political differences within the MDF itself rather than by the abovementioned MDF-SZDSZ “presidential” coalition which, as we already know, gave the green light to the presidential appointment of the SZDSZ-nominee, Árpád Göncz (Deak, 2005:39).

fostered the creation of new political forces, while hindering at the same time the

process of institutionalization of the already existing parties as well as the patterns of

interaction previously established by them. Let´s examine then each of the presidential

elections...and its effects.

The November 1990 presidential elections, won by Solidarity founder and Nobel

laureate Lech Wałęsa in the second round, played a critical role in the definition and

shaping of the structure of inter-party competition in Poland in two different ways: on

the one hand, it reinforced the pattern of fierce opposition between Solidarity and anti-

Solidarity forces already present in the 1989 (semi-free) parliamentary elections; on the

other, it constituted the last straw in the decomposition of the “Solidarity” camp, started

roughly eight months earlier when Wałęsa, then only the leader of the Trade Union,

called for a “war at the top” between the several forces within the movement in order to

facilitate its “orderly” disintegration and the formation of a “Western type” party system

with two different ideological alternatives: left and right.30 Nonetheless, the effect of the

presidential contest on the increase in the number of political parties should not be

exaggerated. At the end of the day, it was the natural consequence of the process of

disintegration of “Forum” type organizations, which also took place in other parts of the

post-communist sphere (Migalski, 2005:55): for example, “Civic Forum” in

Czechoslovakia, “Demos” in Slovenia, etc. However, what can be considered to be a

direct consequence of the confrontation between Wałęsa and Mazowiecki in the first

round was the fragmentation of the forces located in the right side of the political

spectrum, mainly between Christian-democratic/conservative (heirs of those who

supported Lech Wałęsa) and liberal-democratic political parties (heirs of those

supporting Mazowiecki);31 whose interaction, since then and with few exceptions, have

been characterized by a severe political enmity and lack of will to cooperate. On the

contrary, and within the left camp, Bartoszcze´s (PSL) appalling electoral results caused

his substitution for Waldemar Pawlak, a 32 years-old politician without “Solidarity”

past, whose connections with the structures of the ancient regime peasant party [ZLD]

30 Wałęsa´s election also had direct consequences for the stability of the structure of competition as a whole. Thus, considering itself to have been elected by a “true and plebiscitary” majority (Jasiewicz, 1997:134), Wałęsa claimed a direct mandate for his conception of a strong and active presidency with a right to directly intervene in the party system. The fall of Olszewski´s cabinet in June 1992 constitutes the best example (Jasiewicz, 1997:141; Tavits, 2009:97; Krok-Paszkowska, 2001:139). 31 The Democratic Union, which placed an essential role in the development of the Polish party system from 1991 to 2001, came into being in December 1990 from the fusion of the Citizens Movement Democratic Action (ROAD), the Forum of the Democratic Right (FPD) and other advocates of Mazowiecki´s candidature for president.

determined a turn of the party to the left and the beginning of its co-operation with the

post-communist SLD (Migalski, 2005:56). Finally, the 1990 presidential elections had

not other implications in terms of party institutionalization but to spark the formation

and organization of new political forces. As “political parties were still at an early stage

of development” (Sanford, 2002:179), the truth is that it was too early to see any

negative effect on the process of party institutionalization, although the potential for the

latter was already there: let´s think, for example, about the creation of the “X” Party by

Tymińsky immediately after his electoral “success” in the presidential race.

The 1995 and 2000 presidential elections also had important consequences for

the process of party system institutionalization in Poland. Thus, Wałęsa´s defeat against

Kwaśniewski in November 1995 due to the disunity of the forces within the political

right prompted these parties to “temporally” unite in the so-called Solidarity electoral

Action (AWS) in order to contest the parliamentary elections in 1997 (Szczerbiak,

1999), and to collaborate “ephemerally” with the liberal Freedom Union (UW) from

October 1997 up to June 2000. In the same vein, the 2000 presidential elections, where

Kwaśniewski (non-partisan) enjoyed the formal support of his former party (SLD) but

also from the post-Solidarity Labour Union (UP), brought forward the formal

collaboration between these two parties in both an electoral (Millard, 2002:362) and,

later on, governmental alliance in 2001 (see appendix), putting to a certain extend an

end to the so-called “post-communist” cleavage, giving path to a new pattern of inter-

party competition: one based more on economic divisions that on cultural ones.

The two presidential elections here analysed also had an important impact in the

“effective” number of parties. On the one hand, the unexpectedly result of former

premier Olszewski (6.9 percent) constituted the solid basis for the creation of a new

national-populist party, namely: the Reconstruction of Poland Movement – ROP

(Migalski, 2005:61; 2007:219-220). Likewise, definitely thinking in his bid for re-

election two years later, Wałęsa sponsored the creation of a new organization, the Non-

Party Bloc for Support of Reforms in 1993. Although the new party failed to attract

substantial popular support, it managed to reach the 5 percent threshold and enter

parliament.32 On the other, the creation of three new parties in 2001 (PO, PiS, and LPR)

were the immediate consequence of Krzaklewski´s crushing first-round defeat and

32 Wałęsa´s defeat in 1995 also prompted BBWR to join forces with the National-Democratic Party and the National Party in the so-called Bloc for Poland, founded in April 1997 (Migalski, 2007:222).

Olechowski´s strong showing as a runner-up in November 2000 (Jasiewicz, 2007:91,

94).

Notwithstanding what has already been said, it was the process of party

institutionalization which was more unambiguously influenced by the second and third

Polish presidential elections. Thus, the arrival of the BBWR threatened the existence of

other established post-Solidarities parties (e.g., PC or ZChN), whose societal roots were

certainly undermined (Grzybowski and Mikuli, 1998:195). Moreover, and with the

exception of the SLD, during the 1995 presidential race all major partisan organizations

underwent a process of de-institutionalization of some sort.

For some (PSL, UW, UP), the process of candidate selection led to internal

feuds among factions and/or personalities. Others (the Christian nationalist

ZChN), posting no own candidate, feuded about whom they should endorse,

shifting their alliances in the middle of the campaign, which led to a growing

confusion among their own membership, candidates, and voters (Jasiewicz,

1997:162).

Even others were the object of internal splits, like the KPN, where part of its

members regretted the lack of support for Wałęsa already from the first round

(Migalski, 2007:222).

In the same vein, the already difficult process of organizational

institutionalization of AWS, an ideologically heterogeneous formation of more than 30

different political groupings (Szczerbiak, 1999), was definitely brought down by

Krzaklewski´s - the leader of Solidarity at the time and the mastermind behind the

formation of AWS, third place and weak performance in October 2000. The latter

clearly precipitated a major crisis of leadership and identity within the party,

contributing towards its disintegration the year after (Szczerbiak, 2001:105).

Definitively, Olszewski´s association with Krzaklewski´s failed candidacy also

precipitated the disintegration of ROP and the integration of part of its members, headed

by the same Olszewski, into the new hard-right, anti-EU, Catholic-nationalistic LPR

(Millard, 2002). The divisions within AWS were also exploited by Lech Kaczyński, the

increasingly popular Minister of Justice, who together with his twin brother Jarosław

founded a new party (PiS). In the same vein, UW´s failure to nominate a candidate or

straightforwardly support Olechowski´s bid led to internal political struggles which

absolutely mined the unity of the party organization and finally culminated with the

departure of Donald Tusk (and his supporters) to form a new political movement known

as the Civic Platform, together with Olechowski and Maciej Płażyński, former Speaker

of the Sejm (Grzybowski and Mikuli, 1998:201; Migalski, 2005:63-64).

In contrast to the previous elections, the effect of the last presidential contest on

the process of party institutionalization has been relatively weak, although forces like

the Democratic Party (UW´s successor) or the SLD as well as the LPR have suffered,

respectively, from the feebly performance and early withdrawal of their candidates

in/from the presidential race. Conversely, and despite the almost concurrent character of

the parliamentary and presidential elections, the significant increase in the “effective”

number of presidential parties (from 2.9 in the year 2000 to 3.6 in 2005) boosted the

“effective” number of electoral parties in more than one point (from 4.5 in 2001 to 5.9

in 2005). In Elgie´s own words,

[In Poland,] the problem with the majority run-off system was that it

provided few incentives for parties to co-operate in support of a single

presidential candidate. Instead, it encouraged small parties to stand

candidates at the first ballot so as to increase their electoral visibility

(2001:219)

In fact, as pointed earlier, in the event a party´s candidate has a good showing in

the first ballot, it may have some clout in establishing the second round coalitions,

giving him/her the right to demand some compensation for such support later on. This is

what definitely happened in 2005 when both Self-Defence and LPR demanded from

Jarosław Kaczyński, PiS´ leader, their share of governmental power as a compensation

for their support to his twin brother (Lech) during the presidential race.33 The inclusion

of these two anti-establishment-parties (Sokół and Śmigrodzki, 2005:195) in the

parliamentary coalition first,34 and in the cabinet later on, definitely blew up the

previous structure of competition, characterised by the political competition between

two ideologically opposite camps (i.e., post-communist and post-solidarity) and the

permanent exclusion of populist “anti-systemic” forces (Jasiewicz and Jasiewicz-

Betkiewicz, 2006; Szczerbiak, 2007).35

33 A similar offer was made by Self-Defence´s leader Andrzej Lepper to Donal Tusk (PO) between the first and second round of the presidential race. Contrary to PiS´ presidential candidate, the latter decided to refuse any kind of future collaboration between PO and Lepper´s party. 34 PSL, which also supported Kaczyński (Lech) in the second round, also formed part of this “parliamentary” coalition supporting Kazimierz Marcinkiewicz´s minority government. This was the first time a post-communist party supported a right-wing cabinet. 35 In September 2001, needed of a coalition partner in order to secure a governing majority in the Sejm, the SLD considered a coalition with the populist-agrarian party Self-Defence led by Andrzej Lepper, although it was finally rejected in favour of SLD´s old coalition partner instead: the agrarian PSL. Previously, other extreme political forces like KPN or ROP had suffered the same fate.

It follows from all that has been said that semi-presidentialism in Poland has

“provided the potential for new axes of conflict without the mediating effect of long-

established relationship among political parties” (Millard, 2000:59). Moreover, each of

the presidential elections has not only brought a new political realignment in the

political scene, but also foster party fragmentation and de-institutionalization. However,

the low level of systemic institutionalization has not solely derived from the mode of

election of the president but also from the way in which the sequence of presidential

elections has structured partisan competition. In fact, the anticipation of presidential

elections in 1990, 1995, and 2000 undermined the cooperation among the political

forces of the right in 1991 and 1993, fomented their cooperation in 1997, while

fostering its disintegration in 2001, respectively. Even in 2005, when parliamentary

elections took place slightly ahead of presidential ones, the presidential campaign

“contaminated” the former (Szczerbiak, 2007:204), undermining the pre-existing

impetus to Kaczyński´s PiS´ co-operation with Tusk´s PO. In short, semi-

presidentialism has had a negative (direct and indirect) effect on the process of party

system institutionalization in Poland.

Slovakia and Moldova

Slovakia and Moldova constitute, without doubt, the most fascinating cases

among the post-communist Eastern European sample as they represent the two unique

instances of type of regime change. Thus, while the former changed parliamentarism for

semi-presidentialism in 1999, the latter became parliamentarian only in 2000 (before it

constitute a semi-presidential democracy). In this sense, both countries provide a

“natural experiment” in order to examine whether change in the mode of election of the

head of state has any influence in the level of party system institutionalization. Acting

as a real “control” cases, they allow for a comparison between the effects of two

different modes of presidential selection (i.e. direct and indirect) within one country. In

this sense, they helps to keep constant other possible causal variables while, at the same

time, enabling to discover whether the change in the type of regime itself precipitates

change in the level of party system institutionalization. In other words, if the process of

institutionalization respectively declined or increased after the reform of the regime,

then this will tell us something about the pernicious effects of semi-presidentialism on

party system institutionalization. If the contrary is true, then the initial hypotheses will

be dismissed.

The first (indirect) presidential elections36 in Slovakia clearly responded to the

already repeated demand for extraordinary consensus among the political parties, but

not without an exhausting partisan struggle and a tough bargaining process between the

first and the second ballot, when Michal Kováč, a member of the Movement for a

Democratic Slovakia (HZDS) at the time and former speaker of the last Czechoslovak

Federal Assembly, managed to be elected with the support (see appendix) of the junior

coalition partner (SNS) and the main opposition party (SDL´) (Goldman, 1999:62-63;

Malová, 1994:416).37

In a different vein, Slovakia´s 1998 presidential elections as well as Moldova´s

since 2000 clearly fulfilled my second expectation: when not the fruit of a compromise,

the selection of the head of state in parliamentary regimes will simply respond to the

already existing structure of inter-party cooperation and collaboration, with no impact

on the future patterns of partisan interaction. In fact, both the successful Moldovan

presidential elections (2001 and 2005) as well as the disastrous contests in Moldova

(2009) and Slovakia (1998) responded to the previous structure of competition

characterised by the rather polarized confrontation between the parties of the

“communist” (PCRM) and/or “authoritarian” (HDZS and SNS) government against the

parties of the “democratic” opposition.

The PCRM, with a clear majority of 71 seats, easily managed to have his

secretary-general, Vladimir Voronin, elected in April 2001 already during the first

round. Four years later, the PCRM required the votes of the PPCD in order to have

Voronin re-elected. Still, such “ad hoc” support of the latter did not have consequences

in terms of the structure of competition for government, which continued to confront the

governing communist against the rest. The May/June 2009 presidential elections also

responded with the previous pattern, although they diverge in the sense that the ruling

party alone (i.e. PCRM) could not have their candidate elected as it was just one vote

short of the required majority), but it could still block (60 seats) any attempt of the

36 Both the Slovak (1992) and Moldovan (2000) Constitutions required that, in order to be appointed head of state, a candidate needed to obtain a qualified majority of three-fifths of all parliamentary members (90 out of 150 and 61 out of 101, respectively). The main difference is while the former required the celebration of new parliamentary elections with new presidential nominees in case no candidate managed to be elected after two rounds, the latter gives the Moldovan President the possibility to dissolve parliament if such scenario takes place, but only after three rounds. In both cases, however, there is a potential chance for the post to remain vacant, producing an institutional deadlock and a dangerous political crisis (e.g. Slovakia in 1998 or Moldova in 2009). 37 Because the SDL´ made its support conditional on the president´s non-partisan position during his term in office, Kováč immediately suspended his party membership after the election.

opposition to get their candidate elected.38 President Voronin then dissolved the

parliament and new legislative elections took place in July. Unfortunately, the result

was again a deadlock parliament where the PRCM had 48 seats, enough to prevent the

now governing “democratic” coalition to appoint a new president.

A very similar situation took place in Slovakia in 1998 when the ruling coalition

at that time (HZDS, SNS and ZRS) could not have their candidate elected as it has only

82 votes (eight short of the majority required), but Meciar´s HZDS alone (61 seats)

could block any attempt to get a candidate elected without their support (Malová and

Učeň, 1999:503-504). As in Moldova (twice in 2009), the presidency remained vacant.

Although the constitutional gridlock in Slovakia ended with the 1998 legislative

elections, which gave the “democratic” opposition (i.e. SDK, SMK, SDL´ and SOP) the

necessary three-fifths majority needed to reform the constitution and provide for the

popular election of the president (Malová and Lastic, 2001), a solution to the

institutional and political crisis cannot be yet foreseen.39

Contrary to what we have seen, the popular election of the Slovak president did

introduce an element of stability in the structure of competition not even seen at the

time when the President Kovác had no other choice but to intervene in the “normal”

running of the party system (e.g. Moravcik´s cabinet).40 Still, it took some time for

semi-presidentialism to alter the moderated degree of systemic stability achieved in

1998. In fact, the first direct presidential elections held in May 1999, roughly eight

months after the legislative contest, still conformed to the previous dichotomous pattern

of competition: Meciarists versus anti-Meciarists. The timing (quasi-concurrent) and

sequence (parliamentary-presidential) of both legislative and presidential elections

definitively contributed to it.

As soon as both the timing and sequence of the Slovak elections changed (see

table B), presidential elections started to exert their “not-so-mysterious” influence (see

section 2) on the process of systemic institutionalization. Thus, the rapprochement

between nationalistic (SNS and HZDS´ splinter parties) and left-leaning forces (Smer)

in 2006 would have been unthinkable without their joint support to Gašparovič´s 38 A similar scenario took place at the first indirect presidential election when, due to multiple splits and defections within the right-win parties, the PCRM could count on 50 up to 59 votes (Roper, 2008a:122). 39 New parliamentary elections will be held in autumn this year. Moreover, a referendum on the adoption of semi-presidentialism will be held shortly before (September, the 5th). 40Like Havel in the Czech Republic during the 1997 government crisis and Tosovský´s interim care-taker cabinet - which he promoted in the shadows, Kovác´s active interference totally respected the pre-existing government/opposition dichotomy and, therefore, did not cause any damage to the still precarious systemic stability.

candidature in 2004. In my opinion, it was this close electoral collaboration that

favoured the formation of the government populist-nationalistic coalition between

Smer, SNS and HZDS two years later.41 A similar relationship could be established, in

principal, between SDKÚ, SaS, KDH and Most´s electoral support to Radicová

candidature in the last 2009 presidential elections and the formation of the last Slovak

cabinet a few weeks ago.

Figure 6. Type of regime and party system institutionalization in Slovakia and Moldova

Pa

rlia

me

nta

rism

Se

mi-

pre

sid

en

tia

lism

Se

mi-

pre

sid

en

tia

lism

Pa

rlia

me

nta

rism

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

Pa

rty

sy

ste

m i

nst

itu

tio

na

liza

tio

n

(iP

SI)

1993-2001/1994-2000 2001-/2002-2009

Time period

Slovakia Moldova

In a similar vein, the Moldovan 1996 (direct) presidential elections also altered

the existing structure of inter-party competition, pitting the opposition against the

agrarian-bureaucratic PDAM, to a certain extent. In fact, President Lucinschi´s, who ran

as an independent in the 1996 contest, decided to collaborate with his former party

(PDAM) as a response to its support during the second round (EECR, 1997:19; Roper,

2008b:115), and even if the latter did not count anymore with an absolute majority at

the time of Ciubuc´s appointment in January 1997. Moreover, the electoral

confrontation between the President (Snegur), the Premier (Shangeli) and the Speaker of

the Parliament (Lucinski) clearly exacerbated PDAM´s electoral decline from 43.2 in

41 Even in the event of Meciar´s victory in the presidential contest, mainly thanks to the “tacit” support of the governing parties at the time, mainly SDKÚ and ANO (Rybář, 2005:336), the structure of inter-party competition would have been drastically altered as it would have facilitated the rapprochement between HZDS and SDKÚ and its collaboration in an eventual minority government, desired by the two parts (Malová and Rybář, 2008).

1994 to 3.6 in 1998.42 Last but not least, Snegur´s defeat at Lucinschi´s hands clearly

difficulted the governmental collaboration between their respective two supportive

political forces, that is, PMDP and PRDM/CDM (Roper, 2008b).

A quick look at figure 6 above, which quantitatively summarizes the level of

systemic institutionalization for the period before and after the introduction of the

direct/indirect election of the president, confirms the initial hypothesis, namely: party

system institutionalization in parliamentary Slovakia (1993-2001) and Moldova (2001-

2009) is/was much higher than during the semi-presidential period (i.e. Slovakia 2002-

2009; and Moldova, 1994-2000).

Figures 7/8. Change of type of regime in Slovakia and Moldova: “indirect” effects

0

1

2

3

4

5

6

7

1993/4 2000/1 2009

Year

Nu

mb

er

of

ele

cto

ral

pa

rtie

s

(EN

EP

)

Slovakia Moldova

In the same vein, semi-presidentialism has also been indirectly detrimental for

the process of party system institutionalization as it has increased the size of the party

system while hindering, at the same time, the level of party institutionalization. As it

follows from figure 7, which displays the ENEP in three different temporal points (at

the first elections, at the first elections following regime change, at the last elections),

electoral fragmentation increased during semi-presidentialism while reducing at the

time parliamentarism was adopted. The same can be said when the ENEP average for

the two regime type periods is calculated (see table 3). There it can be observed how in

Slovakia the ENEP increased on two points with the introduction of semi-

presidentialism, while decreasing in almost 1.5 points in Moldova after parliamentarism

was adopted. Concrete examples of such negative relationship between the semi-

presidentialism and party system institutionalization are the following: (1) the

42 Similarly, the PCRM´s comeback in 1998 could, most probably, not have been possible without Voronin´s rather successful electoral performance in November 1996 (he came in third place with 10% of the votes).

0

10

20

30

40

50

60

70

80

2000/2001 2009

Year

Pa

rty

Inst

itu

tio

nal

izat

ion

(ip

s)

Slovakia Moldova

presidential ambition of Schuster (SOP),43 Rusko (ANO), Gašparovič (HZD) and, to a

lesser extent, Fico (Smer), clearly contributed to the increasing number of political

forces in the Slovak political spectrum from 1999; (2) similarly, the formation of two

completely new parties at the beginning of the second half of the 1990s (PMDP and

PRCM) was the result of, respectively, Lucinschi´s and Snegur´s bid for the presidency

(Roper, 2008a:119-120).

Table 3. Type of regime, the “effective” number of electoral parties (ENEP) and party institutionalization (PI) in Slovakia and Moldova

Slovakia Moldova Year ENEP (average) PI (ips) Year ENEP (average) PI (ips)

1993-2001 5.5 82.0 1994-2000 4.9 47.9 2002-2009 7.5 81.0 2001-2009 3.5 86.8 Source: Own calculations

But semi-presidentialism in these two countries has also been counterproductive

for the institutionalization of individual political parties themselves. Figure 8 displays

the scores of party stability (ips) in the two countries, both at the moment of regime

change as well as in 2009. In both cases, the level of party institutionalization is clearly

higher at the time of parliamentarism than under semi-presidentialism. Moreover, this

difference can also be perceived when the two regime periods are taken into

consideration as a whole (table 3), especially in Moldova. In this context, it is important

to note that the number of party splits and mergers in Slovakia has been definitively

higher from 1999 than in the previous period (see Rybář, 2004:35).44 In Moldova,

Lucinschi and Shangeli´s electoral struggle for the presidency in 1996 did not but

exacerbate the internal divisions within the once-dominant PDAM. Moreover, as stated

in King (2000:162-163), the Snegur and Lucinschi´s confrontation deprived the later of

the support of his former arch-nemesis Popular Front, which finally merged with

Snegur´s PRCM to form the Democratic Convention of Moldova (DCM).45 All in all,

and in comparison to what can be observed under parliamentarism, the adoption of a

semi-presidential regime negatively affected both party system concentration and party

institutionalization in both Slovakia and Moldova.

In summary, the two previous cases provide a unique proof that change in the

type of regime corresponds to both qualitative and quantitative change in the process of

43 On the presidential ambitions of Rudolf Schuster, SOP´s founder, see Fitzmaurice (2001:323). 44 Interestingly enough, three of the four parties supporting Gašparovič´s bid for presidency in 2004 had split from two previous important party organizations: the Movement from Democracy (HZD) and the People´s Union (LS) from the HZDS, and Smer from the SDL´. 45 Interestingly enough, both parties would merge in 2003 to form the Party Alliance Our Moldova (PAMN).

party system institutionalization, which heavily decreased at the time semi-

presidentialism is in place.

Conclusions

In general, scholars have failed to predict the implications that the type of

regime in general, and semi-presidentialism in particular, can have for the

institutionalization of party systems in young democracies. One obvious reason for this

has been the lack of a consensual definition of semi-presidentialism. Following the most

recent scholarship and for the reasons explained above, I have chosen to adopt Elgie´s

minimal definition of the concept which focusing solely on the mode of election of the

head of state perfectly suits an analysis that avoids at all cost any reference to the

constitutional powers of either the president or the prime minister.

After a brief review of the basic arguments in the type of regime debate, I argued

that two mechanisms account for the weak party system institutionalization in semi-

presidential regimes. The first refers to the mode of election of the president and its

separate electoral legitimacy, which has a direct effect of the structure of inter-party

competition. The logic is that while, on the one hand, the double electoral process,

which requires from any presidential candidate a broader electoral base, many times

across ideological lines, introduces the potential for instability (and unpredictability) in

the patterns of interaction among the different political forces; on the other, the direct

electoral mandate “empowers” the president to intervene in the party system, re-

structuring according to his personal interests, which do not always respond to the

interest of the parties themselves. The second mechanism points to negative effect of

semi-presidentialism on party system fragmentation and party institutionalization,

affecting the level of institutionalization only indirectly. On the one hand, we expect

party system size to vary with the number of presidential parties, and we know from

other previous studies (Casal Bértoa, 2010 and forthcoming): as fragmentation rises, the

structure of competition between parties is likely to face greater instability. On the other

hand, the ability of popular presidential elections to encourage political outsiders as well

as non-party candidates to stand for election, promote partisan factionalism and the

personalization of politics should have a negative impact on the level of party

institutionalization.

A detailed analysis of the mechanisms through which the type of regime affect

party system institutionalization in four post-communist Eastern European countries

reveals that while in Hungary, and Moldova since 2001, parliamentarism not only has

enhanced the stability in the patterns of inter-party competition, but also fostered party

concentration and institutionalization; in Poland and Slovakia (since 2002) semi-

presidentialism has definitely destroyed the pre-existing structure of competition after

every presidential contest. Moreover, the popular election of the president in these two

countries, as well as in Moldova up to 2000, has not only delayed the process of

institutionalization (both at the societal and organizational level) of already existing

political forces, but also increased both the actual and “effective” number of political

parties, in itself boosted by the peculiar institutional arrangements adopted, namely: the

combination of non-concurrent and majority run-off presidential elections.

In sum, and bearing in mind all what has been said, I am on safe ground when

paraphrasing Moestrup to conclude that semi-presidentialism does not appear to be

particularly well-suited for the institutionalization of party systems in young

democracies (2004:228).

References

Amorim Neto, O. and Cox, G. W. (1997): “Electoral Institutions, Cleavage Structures,

and the Number of Parties”, American Journal of Political Science, v. 61, pp. 149-74

Azebedo, E., and Nijzink, L. (2007): “Semi-presidentialism in Guinea-Bissay: the

Lesser of Two Evils?”, in Robert Elgie and Sophia Moestrup (eds.) Semi-

presidentialism Outside Europe. London: Routledge

Bahro, H., Bayerlein, B. H., and Vesser, E. (1998): “Duverger’s Concept: Semi-

Presidential Government Revisited”, European Journal of Political Research, v. 34, pp.

201–24

Bakke, E. and Sitter, N. (2005): “Patterns of Stability. Party Competition and Strategy

in Central Europe since 1989”, Party Politics, v. 11, n. 2, pp. 243-263

Bartolini, S. (1984): “Sistema partitico ed elezione diretta del Capo dello Stato in

Europa”, Revista Italiana di Scienza Politica, v. 14, pp. 223–243

Baylis, T. A. (2007): “Embattled Executives: Prime Ministerial Weakness in East

Central Europe”, Communist and Post-Communist Studies, v. 40, n. 1, pp. 81-106

Bielasiak, Jack. 2002. “The Institutionalization of Electoral and Party Systems in Post-

communist States”, Comparative Politics, v. 34, n. 2, pp. 189-210

Blondel, J., and Müller-Rommel, F. (eds.) (2001): Cabinets in Eastern Europe, New

York: Palgrave Macmillan

Cadoux, C. (2007): “Semi-presidentialism in Madagascar”, in Robert Elgie and Sophia

Moestrup (eds.) Semi-presidentialism Outside Europe. London: Routledge

Casal Bértoa, F. (2010): “A Tale of Four Systems. Party System Institutionalization and

its Sources: the Development of a Concept and its Empirical Application to East Central

European Democracies”, PhD. Dissertation Manuscript

Casal Bértoa, F. (forthcoming): “Sources of Party System Institutionalization in New

Democracies. Lessons from East Central Europe”, EUI Working Paper Series

Casal Bértoa, F. And Enyedi, Z. (2010): “Party System Closure: Conceptualization,

Operationalization, and Validation”, DISC Working Paper Series, n. 11

Casal Bértoa, F., and Mair, P. (forthcoming): “Two decades on: How Institutionalized

are Post-Communist Party Systems?”, EUI Working Paper Series

Colton, T. J. (1995): “Superpresidentialism and Russia’s Backward State”, Post-Soviet

Affairs, v. 11, n. 2, pp. 144-148

Colton, T. J., and C. Skach (2005): “A Fresh Look at Semi-presidentialism. The Russian

Predicament”, Journal of Democracy, v. 16, n. 3, pp. 113-126

Cox, G. (1997): Making Votes Count: Strategic Coordination in the World’s Electoral

Systems. New York: Cambridge University Press

Duverger, M. (1954): Political Parties: Their Organization and Activity in the Modern

State. New York: John Wiley and Sons, Inc

Duverger, M. (1980): “A New Political System Model: Semi-presidential Government,”

European Journal of Political Research, v. 8, pp. 165-187

Duverger, M. (1997): “The Political System of the European Union”, European Journal

of Political Research, v. 31, pp. 125-146

Eastern European Constitutional Review (EECR) (1997), v. 6, pp. 1-123

Elgie, R. (1998): “The Classification of Democratic Regime Types: Conceptual

Ambiguity and Contestable Assumptions”, European Journal of Political Research, v.

33, pp. 219–38

Elgie, R. (ed.) (1999): Semi-Presidentialism in Europe. Oxford: Oxford University

Press

Elgie, R. (2001): “Divided Government in Comparative Perspective”, in Robert Elgie

(ed.) Divided Government in Comparative Perspective. Oxford: Oxford University

Press

Elgie, R. (2004): “Semi-Presidentialism: Concepts, Consequences and Contesting

Explanations”, Political Studies Review, v. 2, pp. 314-330

Elgie, R. (2005): “A Fresh Look at Semi-presidentialism. Variations on a Theme”,

Journal of Democracy, v. 16, n. 3, pp. 98-112

Elgie, R. (2007): “Varieties of Semi-Presidentialism and Their Impact on Nascent

Democracies”, Taiwan Journal of Democracy, v. 3, n. 2, pp. 53-71

Elgie, R. (2008): “The Perils of Semi-presidentialism. Are They Exaggerated?”,

Democratization, v. 15, n. 1, pp. 49-66

Elgie, R., and Moestrup, S. (eds.) (2007): Semi-presidentialism Outside Europe.

London: Routledge

Elgie, R., and Moestrup, S. (eds.) (2008): Semi-presidentialism in Central and Eastern

Europe. Manchester: Manchester University Press

Epstein, L. (1967): Political Parties in Western Democracies. New York: Praeger

Filippov, M., Ordeshook, P., and Shvetsova, O. (1999): “Party Fragmentation and

Presidential Elections in Post-Communist Democracies”, Constitutional Political

Economy, v. 10, pp. 3–26

Fish, M. S. (2001): “The Inner Asian Anomaly: Mongolia’s Democratization in

Comparative Perspective”, Communist and Post-Communist Studies, v. 34, n. 3, pp.

323-338

Fish, M. S. (2001b): “The Executive Deception: Superpresidentialism and the

Degradation of Russian Politics,” in Valerie Sperling (ed.) Building the Russian State:

Institutional Crisis and the Quest for Democratic Governance. Boulder, CO: Westview

Frison-Roche, F. (2005): Le “Modèle semi-présidentiel” comme instrument de la

transition en Europe post-communiste, Bulgarie, Lituanie, Macédoine, Pologne,

Roumanie et Slovénie. Bruxelles: Bruylant

Fitzmaurice, J. (2001): “The Slovak presidential election, May 1999”, Electoral Studies,

v. 20, pp. 321–325

Gallagher, M. (1999): “Republic of Ireland”, in Robert Elgie (ed.), Semi-presidentialism

in Europe. Oxford: Oxford University Press

Golder, M. (2006): “Presidential Coattails and Legislative Fragmentation”, American

Journal of Political Science, v. 50, n. 1, pp. 34–48

Goldman, M. F. (1999): Slovakia since Independence: a Struggle for Democracy. USA:

Praeger Publishers

Grzybowski, M., and Mikuli, P. (1998): “Poland”, in Sten Berglund, Tomas Hellén, and

Frank H. Aarebrot The Handbook of Political Change in Eastern Europe. Cheltenham;

Northampton, MA : Edward Elgar

Grzymała-Busse, A. (2002): Redeeming the Communist Past: The Regeneration of

Communist Successor Parties in East Central Europe After 1989. Cambridge:

Cambridge University Press

Harmel, R. and Janda K. (1982): Parties and their Environments: Limits to Reform?

Longman, New York

Henderson, K. (2002): Slovakia: the Escape from Invisibility, New York: Routledge

Huskey, E. (1997): “Kyrgyzstan: The Fate of Political Liberalization”, in Karen

Dawisha & Bruce Parrott (eds.) Conflict, Cleavage and Change in Central Asia and the

Caucasus. Cambridge: Cambridge University Press

Huskey, E. (2007): “Eurasian Semi-presidentialism: the Development of Kyrgyztan´s

Model of Government”, in Robert Elgie and Sophia Moestrup (eds.) Semi-

presidentialism Outside Europe. London: Routledge

Ilonszki, G., and Kurtán, S. (2006): “Hungary”, European Journal of Political

Research, v. 45, pp. 1120-1127

Ishiyama, J. T., and Kennedy, R. (2001): “Superpresidentialism and Political Party

Development in Russia, Ukraine, Armenia and Kyrgyztan”, Europe-Asia Studies, v. 53,

pp. 1177-1191

Ishiyama, J. T. (2008): “Political Party Development and Party ̒ Gravity̒ in Semi-

Authoritarian States”, Taiwan Journal of Democracy, v. 4, n. 1, pp. 33-53

Jasiewicz, K. (1997): “Wałęsa´s Legacy to the Presidency”, in Raymond Taras (ed.)

Post-communist Presidents. Cambridge: Cambridge University Press

Jasiewicz, K. (2007): “Poland: Party System by Default”, in Paul Webb and Stephen

White (eds.), Party Politics in New Democracies. Oxford: Oxford University Press

Jasiewicz, K., and Jasiewicz-Betkiewicz, A. (2007): “Poland”, European Journal of

Polticial Research, v. 46, pp. 1063-1074

“Citizens and Politics,” chapter 12 in S. White, J. Batt, and P.

Jones, M. P. (1995): Electoral laws and the survival of presidential democracies. Notre

Dame: University of Notre Dame Press

King. C. (2000): The Moldovans: Romania, Russia, and the Politics of Culture,

Stanford: Hoover Institution Press

Kirschke, L. (2007): “Semi-presidentialism and the Perils of Power-Sharing in

Neopatrimonial States”, Comparative Political Studies, v. 40, n. 11, pp. 1372-1394

Krok-Paszkowska, A. (2001): “Divided Government in Poland”, in Robert Elgie (ed.)

Divided Government in Comparative Perspective. Oxford: Oxford University Press

Kuenzi, M., and Lambright, G. (2001): “Party System Institutionalization in 30 African

Countries”, Party Politics, v. 7, n. 4, pp. 437-468

Lewis, P. G. (2000): Political Parties in Post-communist Eastern Europe. London:

Routledge

Lindberg, S.I. (2007): “Institutionalization of Party Systems? Stability and Fluidity

among Legislative Parties in Africa´s Democracies”, Government and Opposition, v.

42, n. 2, pp. 215-241

Linz, J. (1990a): “The Perils of Presidentialism”, Journal of Democracy, v. 1, pp. 51-69

Linz, J. (1990b): “The Virtues of Parliamentarism”, Journal of Democracy, v. 1, pp. 84-

91

Linz, J. (1994): “Presidential versus Parliamentary Democracy: Does it Make a

Difference?”, in Juan Linz and Arturo Valenzuela (eds.), The Failure of Presidential

Democracy. Baltimore: Johns Hopkins University Press

Lijphart, A. (1994): Electoral Systems and Party Systems: A Study of Twenty-Seven

Democracies 1945-1990, Oxford: Oxford University Press

Lijphart, A. (2004); “Constitutional design for divided societies”, Journal of

Democracy, v. 15, n. 2, pp. 96-109

Mainwaring, S. (1993): “The Presidentialism, Multipartism, and Democracy: The

Difficult Combination”, Comparative Political Studies, v. 26, n. 2, pp. 198–228

Mainwaring, S. (1998): “Party Systems in the Third Wave”, Journal of Democracy, v.

9, n. 3, pp. 67-81

Mainwaring, S., and Scully, T. (1995): Building Democratic Institutions: Party Systems

in Latin America. Stanford, CA: Stanford University Press

Mainwaring, S., and Shugart, M. S. (1997): Presidentialism and Democracy in Latin

America. New York: Cambridge University Press

Mair, P. (1996): “Comparing Party Systems”, in Lawrence LeDuc, Richard G. Niemi,

and Pippa Norris (eds.) Comparing Democracies: Elections and Voting in Global

Perspectives, London: Sage

Mair, P. (1997): Party System Change. Approaches and Interpretations, Oxford:

Clarendon Press

Mair, P. (2001): “The Freezing Hypothesis: an Evaluation”, in Lauri Karvonen and

Stein Kuhnle (eds.), Party Systems and Voter Alignments Revisited, London: Routledge

Mair, P. (2006): “Party System Change”, in Richard S. Katz and William Crotty (eds.),

Handbook of political parties, London: Sage

Mair, P. (2007): “Party Systems and Alternation in Government, 1950-2000: Innovation

and Institutionalization”, in Siri Gloppen and Lise Rakner (eds.) Globalisation and

Democratisation: Challenges for Political Parties, Bergen: Fagbokforlaget

Malová, D. (1994): “Slovakia”, European Journal of Political Research, v. 26, n. 3-4,

pp. 413-421

Malová, D., and Láštic, E. (2001): The gradual amending of the Slovak constitution:

Combating the ambiguous rules in 1992–2001. Central European Political Science

Review, v. 2, n. 4, pp. 103–127

Malová, D., and Učeň, P. (1999): “Slovakia”, European Journal of Political Research,

v. 36, pp. 497-506

Meleshevich, A. (2007): Party Systems in Post-Soviet Countries: a Comparative Study

of Political Institutionalization in the Baltic States, Russia, and Ukraine, New York:

Palgrave

Migalski, M. (ed.) (2005): Parties and party systems in Central and Eastern Europe.

Sosnowiec: Wyższa Szkoła Zarządzania i Marketingu

Migalski, M. (2007): “Polski System Partyjny”, in Rafał Glajcar and Marek Migalski

(ed.) Prezydent w Polsce po 1989 r. studium politologiczne. Wydawnictwo: Sejmowe

Millard, F. (2000): “Presidents and Democratization in Poland: the Roles of Lech

Wałęsa and Aleksander Kwaśniewski in Building a New Polity”, Journal of Communist

Studies and Transition Politics, v. 16, n. 3, pp. 39-62

Millard, F. (2002): “The Presidential Election in Poland, October 2000”, Electoral

Studies, v. 21, pp. 357-63

Moestrup, S. (2004): Semi-presidentialism in Comparative Perspective: Its Effects on

Democratic Survival. PhD Diss.: George Washington University

Morlino, L. (1998): Democracy between Consolidation and Crisis: Parties, Groups,

and Citizens in Southern Europe. Oxford: Oxford University Press

Moser, Robert (1998): “The Electoral Effects of Presidentialism in Post-Soviet Russia”,

in John Lowenhardt (ed.), Party Politics in Post-Communist Russia. London: Frank

Casso

Moser, R. (2001): Unexpected outcomes: electoral systems, political parties, and

representation in Russia. Pittsburgh: University of Pittsburgh Press

Mozaffar, Shaheen, Scarritt, J. R., and Galaich, G. (2003): “Electoral Institutions,

Ethnopolitical Cleavages and Party Systems in Africa’s Emerging Democracies.”

American Political Science Review, v. 97, n. 3, pp. 379–90

O’Neil, P. (1993): “Presidential Power in Post-Communist Europe: The Hungarian Case

in Comparative Perspective”, Journal of Communist Studies, v. 9, n. 3, pp. 177-201

O´Neil, P. (1997): “Hungary: Political Transition and Executive Conflict: the Balance

or Fragmentation of Power?, in Raymond Taras (ed.) Post-communist Presidents.

Cambridge: Cambridge University Press

Pasquino, G. (1997): “Semi-Presidentialism: A Political Model at Work”, European

Journal of Political Research, v. 31, pp. 128–37

Protsyk, O. (2005a): “Politics of Intra-executive Conflict in Semi-presidential Regimes

in Eastern Europe,” East European Politics and Society, v. 18, n. 2, pp. 1-23

Protsyk, O. (2005b): “Prime Ministers´ Identity in Semi-Presidential Regimes:

Constitutional Norms and Cabinet Formation Outcomes”, European Journal of Political

Research, v. 44, pp. 721-748

Przeworski, A. (1975): “Institutionalization of Voting Patterns, or is Mobilization the

Source of Decay?, American Political Science Review, v. 69, n. 1, pp. 49-67

Roper, S. D. (2002): “Are All Semipresidential Regimes the Same? A Comparison of

Premier-Presidential Regimes”, Comparative Politics, v. 34, n. 3, pp. 253–72

Roper, S.D. (2008a): “From Semi-presidentialism to Parliamentarism: Regime Change

and Presidential Power in Moldova”, Europe-Asia Studies, v. 60, n. 1, pp. 113-126

Roper, S.D. (2008b): “The Impact of Party Fragmentation on Moldovan Semi-

presidentialism”, in Robert Elgie and Sophia Moestrup (eds.) Semi-presidentialism in

Central and Eastern Europe, Manchester: Manchester University Press

Rose, R., and Tikhomirov, E. (1996): “Russia's Forced-Choice Presidential Election”,

Post-Soviet Affairs, v. 12, n. 4, pp.351-379

Rybář, M. (2005): “The 2009 presidential election in Slovakia”, Electoral Studies, v.

24, pp. 303–344

Rybář, M. (forthcoming): “The Presidential Election in Slovakia”, Electoral Studies

Sanford, G. (2002): Democratic Government in Poland. Palgrave Macmillan

Samuels, D. J. (2000): “The Gubernatorial Coattails Effect: Federalism and

Congressional Elections in Brazil”, Journal of Politics, v. 62, n. 1, pp. 240–53

Samuels, D. J. (2002): “Presidentialized Parties. The Separation of Power and Party

Organization and Behavior”, Comparative Political Studies, v. 35, n. 4, pp. 461-483

Sartori, G. (1976): Parties and Party Systems. A Framework for Analysis, Volume I,

Cambridge: Cambridge University Press

Sartori, G. (1997): Comparative Constitutional Engineering: An Inquiry into Structures,

Incentives and Outcomes. London: Macmillan

Schleiter, P., and Morgan-Jones, E. (2006): “President, Assembly and Cabinet

Composition in European Semi-Presidential Democracies”, Paper presented at the

Annual Meeting of the American Political Science Association (Philadelphia, 31

August)

Sedelius, T. (2006): The Tug-of-War between Presidents and Prime Ministers. Semi-

Presidentialism in Central and Eastern Europe. PhD Diss.: Örebro University, Sweden

Shin-Goo, K. (2004): “The Influence of Presidential Heads of State on Government

Formation in European Democracies: Empirical Evidences”, Paper presented at the

Annual Meeting of the Midwest Political Science Association (Chicago, 15-18 April)

Shugart, M. S. (2005): “Semi-presidential Systems: Dual Executive and Mixed

Authority Patterns,” French Politics, v. 3, n. 3, pp. 323-351

Shugart, M. S. (2006): “Comparative Executive-Legislative Relations”, in R.A.W.

Rhodes, Sarah A. Binder and Bert A. Rockman (eds.), The Oxford Handbook of

Political Institutions, Oxford: Oxford University Press

Shugart, M. S., and Carey, J. M. (1992): Presidents and Assemblies: Constitutional

Design and Electoral Dynamics. Cambridge: Cambridge University Press

Skach, C. (2005): Borrowing Constitutional Designs: Constitutional Law in Weimar

Germany and the French Fifth Republic. Princeton, NJ: Princeton University Press

Sokół, W., and Śmigrodzki (eds.) (2005): Współczesne Partie i Systemy Partyjny.

Zagadnienia teorii i praktyki politycznej. Lublin: UMCS

Spörer, D. (2004): “Ausmass, Strukturen und Implikationen präsidialer Macht im

postkommunistischen Raum”, Swiss Political Science Review, v. 10, pp. 157-179

Stepan, A. and Skach, C. (1993): “Constitutional Frameworks and Democratic

Consolidation: Parliamentarianism versus Presidentialism”, World Politics, v. 46, n. 2,

pp. 1-22

Stepan, A., and Suleiman, E. N. (1995): “The French Fifth Republic: A model for

import? Reflections on Poland and Brazil”, in H. E. Chehabi & Alfred Stepan (eds.),

Politics, Society, and Democracy. Comparative Studies, Boulder: Westview Press

Stockton, H. (2001): “Political Parties, Party Systems, and Democracy in East Asia:

Lessons from Latin America”, Comparative Political Studies, v. 34, n.1, pp. 94-119

Szczerbiak, A. (1999): “Interests and Values: Polish Political Parties and their

Electorates”, Europe-Asia Studies, v. 51, n. 8, pp. 1401-1432

Szczerbiak, A. (2001): “Explaining Kwasniewski's Landslide: The October 2000 Polish

Presidential Election”, Journal of Communist Studies and Transition Politics, v. 17, n.

4, pp. 78-107

Szczerbiak, A. (2007): “̒Social Poland̒ Defeats ʻLiberal Poland̒ ? The

September/October 2005 Polish Parliamentary and Presidential Elections”, Journal of

Communist Studies and Transition Politics, v. 23, n. 2, pp. 203-232

Tavits, M. (2009): Presidents with Prime Ministers. Oxford: Oxford University Press

Toole, J. (2000): “Government Formation and Party System Stabilization in East

Central Europe”, Party Politics, v. 6, n. 4, pp. 441-461

Učeň, P. (2005): “Slovakia”, European Journal of Political Research, v. 44, pp. 1167-

1178

Valenzuela, A. (2004): “Latin American presidencies interrupted”, Journal of

Democracy, v. 15, n. 4, pp. 5-19

White Stephen, Wyman, M., and Kryshtanovskaya, O. (1995): “Parties and Politics in

Post-Communist Russia”, Communist and Post-Communist Studies, v. 25, pp. 185-202

Wright, S., and Riker, W. (1989) “Plurality and Runoff Systems and Numbers of

Candidates”, Public Choice, v. 60, n. 2, pp. 155-175

APPENDIX Table A. Effects (direct and indirect) of the type of regime on the process of party system institutionalization in post-communist Eastern Europe

Country iPSI ENEP PI (“ips”) Parliamentary regimes (n = 8)

Kosovo 4.2 4.8 n/a Hungary 2.1 5.2 90.8 Albania 1.5 4.1 91.5 Czech R. 1.5 5.4 86.0

Moldova (2001-) 0.9 3.5 86.8 Slovakia (-2001) 0.7 5.5 82.0

Latvia -0.5 7.4 73.9 Estonia -0.9 5.8 81.7 TOTAL 1.2 5.2 84.7

Semi-presidential regimes (n =11)*

Croatia 1.3 4.4 88.9 Slovenia 1.0 6.1 84.6

Moldova (-2000) 0.7 4.9 47.9 Serbia 0.5 4.5 78.1

Slovakia (2002-) 0.1 7.5 81.0 Macedonia 0 4.2 79.4 Romania 0 5.2 85.4 Bulgaria -0.5 4.2 79.7 Lithuania -1.0 7.0 75.2 Poland -1.9 7.0 72.3 Ukraine -9.7 18.3 58.7

Estonia (1992) - 8.9 - TOTAL -0.9 6.7 75.6

* Direct presidential elections were held in Estonia in January 1992.

Table B. Parliamentary and presidential (s)elections/coalitions in Hungary, Poland, Slovakia and Moldova (1990-2009) Parliamentary

elections Government

parties Presidential (s)election

President “Presidential” coalition*

Hungary March/April

1990 MDF-FKGP-

KNDP May 1990 Árpad

Göncz MDF-SZDSZ

May 1994 MSDZ-SZDSZ June 1995 Árpad Göncz MSZP-SZDSZ May 1998 Fidesz-MDF-

FKGP June 2000 Ferenc

Mádl Fidesz-FKGP

April 2002 MSZP-SZDSZ April 2006 MSZP-SZDSZ

August 2005 Laszlo Sólyom

Fidesz-MDF

Poland October

1991 WAK(ZChN)

-PC-PL

September 1993

SLD-PSL

November/ December

1990

Lech

Wałęsa

PC-ZChN-PL-

KLD-etc.

September 1997

AWS-UW November 1995

Aleksander Kwaśniewski

SLD

September 2001

SLD/UP-PSL October 2000

Aleksander Kwaśniewski

SLD46-UP

September 2005

PiS October 2005

Lech Kaczyński

PiS-LPR

October 2007 PO-PSL June 2010 Bronisław Komorowski

PO47

Slovakia (indirect election)48 June 1992 HZDS-SNS September/

October 1994 HZDS-SNS-

ZRS September

1998 SDK-SDL´-SMK-SOP

February 1993

Michal Kováč

HZDS (-SNS-SDL´)

Slovakia (direct election) May 1999 Rudolf

Schuster SDK-SDL´-SMK-

SOP

September 2002

SDKU-KDH-SMK-ANO April 2004 Ivan

Gašparovič HZD-LS

(-SNS-Smer) June 2006 Smer-HZDS-

SNS March/April

2009 Ivan

Gašparovič HZD49-Smer-

SNS * The party of the president is underlined.

46 In 2000, Aleksander Kwaśniewski, although a former member of the SLD which supported him once again, was not the Alliance´s formal candidate. 47 Although the SLD did not officially support Komorowski in the second round, some of its more important leaders (e.g. former secretary-general Wojciech Olejniczak) did. 48 Similarly, the National Council of the Slovak Republic was considered the legal successor of the Slovak National Council within the Czechoslovak Federal Republic. 49 In 2009, Ivan Gašparovič, although a former member of the HZD which supported him once again, was not a member of the party.

Table B. (continuation) Parliamentary

elections Government

parties Presidential (s)election

President “Presidential” coalition*

Moldova (direct election) April 1994 PDAM May 1998 CDM-PMDP-

PFD

Nov./Dec. 1996

Petru Lucinschi

PDAM50-PSMUE

Moldova (indirect election) April 2001 PCRM April 2001 Vladimir

Voronin PCRM

April 2005 PRCM April 2005 Vladimir Voronin

PCRM (-PPCD-PSL-

PDM) April 2009 PCRM May/June 2009 Vladimir

Voronin (“acting”)

-

Sept. 2009 PLDM-PL-PDM-PAMN

Nov. 2009 Mihai Ghimpu (acting)

-

* The party of the president is underlined.

Political parties: Hungary: Fidesz = Federation of Young Democrats/Hungarian Civic Party; FKGP = Independent Party of Smallholders; KNDP = Christian Democratic People´s Party; MDP = Hungarian Democratic Forum; MSZD = Hungarian Socialist Party; SZDSZ = Alliance of Free Democrats; Poland: AWS = Solidarity Electoral Action; KLD = Liberal Democratic Congress; LPR = League of Polish Families; PC = Centre Alliance; PChD = Party of Christian Democrats; PiS = Law and Justice; PL = Peasant Alliance; PO = Civic Platform; PPG = Polish Economic Programme; PSL = Polish Peasant Party; SDPL = Social Democracy of Poland; SLCh = Christian People´s Party; SLD = Democratic Left Alliance; SO = Self-Defense; UD = Democratic Union; UP = Union of Labour; UW = Freedom Union; WAK= Catholic Electoral Action; Slovakia: ADSR = Alliance of Democrats of the Slovak Republic; ANO = Alliance of the New Citizen; APR = Alternative of Political Realism; HZDS = Movement for a Democratic Slovakia; KDH = Christian Democratic Movement; Most = Bridge; NDK = National Democratic Party; SaS = Freedom and Solidarity; SDL´= Party of the Democratic Left; SDK = Slovak Democratic Coalition; SDKÚ = Slovak Democratic and Christian Union; Smer = Direction/Social Democracy; SMK = Hungarian Coalition; SNS = Slovak National Party; SOP = Party of Civic Understanding; ZRS = Association of Workers of Slovakia.; Moldova: CDM = Democratic Convention of Moldova; PAMN = Our Moldova Alliance; PCRM = Party of Communists of the Republic of Moldova; PDAM = Democratic Agrarian Party of Moldova; PFD= Party of Democratic Forces; PL = Liberal Party; PLDM= Liberal Democratic Party of Moldova; PDM (PMDP) = Democratic Party of Moldova (For a Democratic and Prosperous Moldova); PSMUE = Socialist Party and “Unitate-Edinstvo” Movement Bloc.

50 Although a former member, the Moldovan Democratic Agrarian Party did not support Lucinski during the first round, but Andrei Shangeli, Moldova´s prime minister at the time.